Improving Laboratory Performance Through Quality Control - Five Simple Steps for QC Success. It is easy to get caught up in an abundance of QC statistics and forget the fundamental reason why QC exists in the first instance. QC is about detecting errors and ensuring that the results you produce are accurate and reliable. All QC procedures should focus on reducing the risk of harm to the patient. We are not examining statistics; we are examining real patients, real results and real lives. Around 70% of all medical decisions are based on laboratory results, which is why it is of utmost importance that each and every laboratory, has a well-designed QC procedure in place.

1. Q
QUALITY CONTROL
Designing an Appropriate QC Procedure for Your Laboratory
Complete QC solutions for results you can trust
Improving Laboratory Performance Through Quality Control
Five simple steps for QC success

2. 2
1
5
The Importance of a well-designed QC Procedure
It is easy to get caught up in an abundance of QC statistics and forget the fundamental reason why
QC exists in the first instance. QC is about detecting errors and ensuring that the results you
produce are accurate and reliable. All QC procedures should focus on reducing the risk of harm
to the patient.We are not examining statistics; we are examining real patients, real results and real
lives. Around 70% of all medical decisions are based on laboratory results, which is why it is of
utmost importance that each and every laboratory, has a well-designed QC procedure in place.
An effective QC strategy is not as complicated as you might think, it does not
require you to become an ‘advanced mathematician’. If you stick to these five simple
steps you can be assured that you are releasing quality QC results.
2
5 Simple Steps
for QC Success
3
Identify the Quality
Specifications for the test
Choose Good Quality
Control Materials
4
Start and End
Patient testing with
a QC evaluation
Ensure you know
the characteristics
of good QC results
Ensure you are
able to recognise
and troubleshoot
an out-of-control
event

3. 3
1. Identify the Quality Specifications for the test
• Not every test performs the same
• Not every test needs to perform the same
• Give poor-performers and high risk tests more QC attention
Not every test in the laboratory performs the same way. Some tests for example are the well-known, well liked, laboratory “good-
performers, always showing excellent recovery rates. However, other tests such as the laboratory “poor-performers”, show a lot
more inconsistencies in performance, and can often be slightly more problematic and frustrating.
You can identify which tests generally perform better, through monitoring your precision over time, via method validation and
comparing your performance to peers through either an EQA scheme or through a peer-group reporting software. Using this
differentiation, afford your“poor-performer” QC tests, a little more QC attention. Run QC more frequently and monitor the results
closely, to ensure that they are both accurate and reliable.
It is also important to remember that not every test needs to perform the same. Some tests, compared to others, have a little more
scope for error before they have an impact on patient care. It is therefore important to ensure that you identify the important
“high risk tests”. Any tests that have one or more of the following characteristics, should be considered high risk:
• A test where there could be a detrimental consequence, should the wrong result be released
• A test that supports the clinicians decision in isolation
• A test that is acted upon immediately
• A test that is performed on a specimen that is difficult/painful to collect
Additionally, make sure you take note of the CLIA analytical quality requirements, to identify the quality specifications of particular
tests. For any test that you identify as high risk, make sure you run QC more frequently and play close attention to your QC results.
“It’s important that you know which tests are your well performing tests
and which are your poor performing tests.”

4. 4
2. Choose Good Quality Control Materials
The effectiveness of any QC procedure is underpinned by the quality of the control materials that you choose to use. ISO 15189
recommends that the
It is important to note that differently manufactured QC material, will not perform in the same way. It is of utmost importance that
you choose the best quality material available on the market, as you may otherwise be putting patients at risk.
Look out for the following qualities in your QC material:
• Have accurately assigned values; the narrower the control ranges used, the more efficient you will be at assessing your
performance. Unfortunately, many manufacturers assign their QC values based on only a handful of results, meaning that control
ranges are far too wide to monitor performance effectively. Choose materials that have assigned values based on thousands of
laboratory results.
• QC material that is “as close as possible to a patient samples”; as recommended by ISO 15189; In order to behave
and react like a patient sample, the material needs to be in a matrix similar to a patient’s as this will allow the best assessment of
performance. Look out for materials that are described as 100% or fully human and look out for lyophilised products that are free
from any artificial materials and preservatives. Many manufacturers describe their material as “human-based”, as these materials
contain components of animal origin and will not be “as close to a patient sample” as possible. This is particularly important
for anti-body based tests. Make sure you check your QC kit insert to ensure you are using 100% human material and that your
lyophilised material is free from preservatives.
• A format suitable for use; QC materials come in lyophilised, liquid-frozen and liquid ready-to-use formats. Lyophilised materials
are great for hotter climates. Liquid frozen and liquid ready-to-use samples, eliminate the chance of reconstitution errors, with
liquid ready-to-use samples requiring the least effort for preparation, saving you time and making them
ideal for point of care testing.
• Are they third party materials; ISO 15189 recommends the “use of
independent third party control materials”.Third party materials offer many
advantages, enabling an unbiased assessment of performance that can be
used on all instrument platforms.
• Cover the full clinical range; ISO 15189 recommends that the
“laboratory should choose concentrations of control materials,
wherever possible, especially at or near clinical decision values,
which ensure the validity of the decisions made”.
• Match their stability claims; Many manufacturers, in an
attempt to make their material ‘look better’, give inaccurate
stability claims and often give different stability claims for different
analytes in the material. Don’t forget that a QC material is only as
stable as its least stable analyte. Make sure your QC manufacturer
is being honest about their stability claims and avoid those who
make different stability claims for different analytes in the same
sample.
“laboratory shall use quality control materials that react to the
examining system in a manner as close as possible to patient samples”.

5. 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
+3
+2
+1
M
-1s
-2s
-3s
+3
+2
+1
M
-1s
-2s
-3s
68.26%
95.46%
99.73%
3. Start and End Patient testing with a QC evaluation
Make sure you are testing patient samples in “batches” and are starting and ending patient sampling with a QC evaluation. If you
only run QC every morning and one morning you find that your QC is out of range, you may be unaware as to whether the
malfunction occurred the previous day or not.Your staff will have to troubleshoot, without knowing if the previous days results have
been affected or not, by re-running samples, perhaps unnecessarily.
Running QC at the end of the day, assures you that that day’s results are not affected, saving you a lot of time and money! If your
lab runs in 24hr operation, make the time between QC evaluations shorter than the time needed to correct erroneous results, as
this is the most cost effective strategy for your lab. Additionally, ensure that you know the number of patient results between QC
evaluations, as this will help you ascertain the number of patients at risk, should an out-of control event occur.
4. Ensure you know the characteristics of good QC results
In order to recognise a “bad” result, it is important that you are able to recognise the qualities of a “good” set of QC results. In an
ideal world, the characteristics of good/normal QC results on a Levey-Jennings chart should;
• Fluctuate randomly around the mean for comparison. There should be an equal number of results above the mean
as below the mean.
• Very rarely exceed 3SD; 3 in every 1000 results should be outside 3SD
• On the whole be within 1SD.
• On Occasion exceed 2SD; It is normal for 1/20 results to exceed 2SD
It is important to keep these normal characteristics in mind when examining your QC results, to ensure you are not falsely rejecting
acceptable results. Ensure you are maintaining a record of your QC results, to evaluate over time, the norm for your laboratory.
There are various QC management software packages available on the market, that make monitoring your QC results easy. Some
QC management software packages also have functionality that allows you to compare your results to other laboratories, using the
same system as yourself. Knowing that you are comparing well to others, provides you with confidence that your results are in fact
good/normal. Look out for software that allows:
• Multiple instrument registrations
• Results to be entered online, anywhere, any time.
• Real time monitoring of peer group data. Ideally peer group data should be updated daily.
• Interactive charts such as Levey-Jennings and Histogram charts
• Different user level accounts so that lab managers can keep trace of the entering and acceptance of results.
• User defined multi-rules, to be applied to QC results, so that results can automatically be rejected or accepted.

6. 6
5. Ensure you are able to recognise and troubleshoot an out-of-control event
• The most effective way to recognise errors is to use QC multi-rules
• Participate in a ISO 17043 accredited EQA scheme to help recognise errors
• Make sure you estimate the magnitude and size of the out of control event, before you correct it
There are various approaches for recognising possible out-of-control events. Some labs use a single rule and other labs opt to use
QC multi-rules. It is important for whatever rule(s) you use, that you are able to identify errors quickly and effectively, without
falsely rejected results, wasting time and money. It is also important that you choose assayed QC material that has tight ranges or
conversely assign the ranges yourself to ensure you have effective error detection.
Recognising out-of-control QC results using a Single rule
A laboratory favourite is the 1 result outside 2SD rule.This is a great rule for alarming possible out-of-control events, however, you
must be careful as it does have a high false rejection rate. Remember, it is normal for 1/20 results to be outside 2SD! It is important
that you also consider the strategy you use when you have a result outside 2SD. Avoid the “repeat, repeat, repeat… got lucky!”
tactic.This is not an effective error detection method, as if 2/3 results exceed 2SD you may have a problem at hand that is worth
investigating further. In short, be careful when using 1 result outside 2SD rule and ensure that if you do repeat, you do so only
once!
Another single rule that is sometimes used, is when 1 result is outside 3SD.This rule has a very low false rejection rate. Remember
only 3 in 1000 results should be outside 3SD. However, this is not the best rule for sensitive error detection. This rule ideally
shouldn’t be used in isolation.
Recognising out-of-control QC results using Multi-rules
Using a combination of multi-rules is the most effective way of recognising out-of-control events. Applying multi-rules means that
you will have a high rate of error detection, whilst maintaining a low rate of false rejection.This means you will run less unnecessary
repeats and waste less time carrying out unnecessary troubleshooting, in turn saving you money.
As a general rule, apply more multi-rules to poor-performer tests and high risk tests.With stable, good-performer, you can use less
multi-rules.The diagram below summarises a variety of commonly used multi-rules.
Control
Data
1
2S
1
3S
2
2S
R
2s
4
1S
10
X
No
Yes
IN - CONTROL ACCEPT RUN
OUT - OF - CONTROL REJECT RUN
No No No No
Yes Yes Yes Yes Yes

7. 7
Conclusion
Armed with these 5 simple steps, you can be assured that you are on the way to producing
an effective and simple QC strategy, reducing the risk of harm to your patients. Remember
that you do not need to become a statistician to design effective QC procedures.
Improvement in your QC procedures simply requires the desire and determination to
take incremental steps. With each step you take, the next step becomes easier and you
will soon reach your goal of QC success!
When an out-of-control event has occurred, ISO 15189 requires laboratories to
“evaluate the results from patient samples that were examined
after the last QC event”.
Participate in an Accredited EQA Scheme
QC alone isn’t sufficient for detecting errors and so you should participate in an IS0 17043 accredited EQA scheme. ISO 15189
states that the “laboratory should participate in interlaboratory comparison programmes”. Not only will EQA help you detect
errors, it will provide you, the physicians and patients you supply results to, confidence and evidence that you are releasing accurate
patient results.
Troubleshooting Out-of-Control Events
Make sure you estimate the magnitude and size of the out of control event before you correct it. It’s a good idea to take note and
monitor your average patient mean or test a known patient sample.That way, you can measure the extent of the problem and the
effect the out-of-control event has had on patient results. Measuring the direction and magnitude of the shift in results, can help you
decide whether any clinical significant errors may have occurred and whether or not you need to repeat the patient results.
Ensure you know how many samples were run from the last QC event and do not release any patient results until the problem has
been rectified.

8. 8
Information correct at time of print. Randox Laboratories Limited is a subsidiary of Randox Holdings Limited a company registered within Northern Ireland with company number N.I. 614690. VAT Registered
Number: GB 151 6827 08. Product availability may vary from country to country. Please contact your local Randox representative for information. Products may be for Research Use Only and not for use in diagnostic
procedures in the USA.
QUALITY CONTROL
Randox Laboratories Limited, 55 Diamond Road, Crumlin, County Antrim, BT29 4QY, United Kingdom
+44 (0) 28 9442 2413 +44 (0) 28 9445 2912 marketing@randox.com randoxqc.com
find out more
24•7
Compatible for use with the Acusera range of third party controls, the Acusera
24•7 software is designed to help laboratories monitor and interpret their QC data.
Access to an impressive range of features including interactive charts and real time
peer group data generated from our extensive database of laboratory participants,
ensures Acusera 24•7 is the most comprehensive package available.
Online QC software with real-time peer group statistics
Comprising over 360 routine and esoteric parameters in 24 comprehensive and
flexible EQA programmes, RIQAS is designed to cover all areas of clinical testing.
Each programme benefits from a wide range of concentrations, frequent reporting
and comprehensive yet users friendly reports.
The largest global EQA scheme with over 31,000 lab participants
Uniquely combining more than 100 parameters conveniently in a single control,
laboratories can significantly reduce costs and consolidate without compromising
on quality. As true third party controls, unbiased performance assessment with any
instrument or method is guaranteed.
True third party controls offering complete test menu consolidation
MAR15